As Budapest’s unified transport authority, BKK manages the public transit, shared...

It's not clear whether Virgin Galactic has the cash reserves to fund a prolonged test phase.

The issue isn't artificial intelligence, but rather an industry adding AI agent integrations into production environments before proper security testing.

See what you missed in Daily Tech Insider from April 27–May 1. The post AI Power Plays, Security Breaches, and Industry Shifts Define the Week in Tech appeared first on TechRepublic.

Elon Musk spent the better part of three days on the witness stand this week in his lawsuit against OpenAI, and it’s already getting messy. Emails, texts, and his own tweets are surfacing in court, and there are plenty more witnesses to come. Musk’s argument against OpenAI? By converting the company to a for-profit model, Sam Altman betrayed the “nonprofit for the […]

buritora/Shutterstock A century ago, the five-day working week helped reshape society. It was introduced at scale by industrial pioneers to address not only worker wellbeing but also economic pressures. US industrialist Henry Ford was among the first to give workers two full days off per week, 100 years ago this month. Ford suspected that giving workers a “weekend” would increase overall productivity – and he was correct. Today, as advances in artificial intelligence accelerate and concerns about job security grow, a similar question is emerging. Could reducing working time again help societies adapt to these seismic changes? The evidence increasingly suggests it can, but not in the simplistic way that is often portrayed. The four-day week is not just a workplace benefit. It is a potential tool to improve wellbeing, support families and rethink how work is distributed in society. Research across multiple countries, including large-scale pilots in the UK and Portugal, shows that reducing working time can deliver meaningful benefits for both employees and organisations. In a 2025 study of four-day week adoption, my colleagues and I found improvements in sleep, exercise and quality of working life. There were positive implications for both the mental and physical health of employees. Our research showed productivity at work can also increase, alongside reductions in absenteeism and staff turnover. And it can be beneficial for an employer’s social image. However, the most important insight is not about productivity but what happens outside work. After all, time is a social resource, not just an economic one. When people move to a four-day week, they do not simply rest more. They reallocate time in ways that have broader implications for society. Across our research, participants said they spend more time with family and friends, engaging in community activities and investing in their physical and mental health by exercising and practising hobbies and self-care activities. These are not trivial changes. Over time, they contribute to stronger social ties, better mental health and more resilient communities. There are also important gender implications. Early findings suggest that reduced working time can lead to fathers being more involved in caring for their children and other domestic responsibilities. While this does not automatically solve gender inequality, it creates conditions that make more equal divisions of labour possible. In this sense, the four-day week is not just about work. It is about how societies organise care, relationships and everyday life. The challenge in service sectors Critics of a four-day week often point out that it is harder to implement in service sectors such as healthcare, childcare, manufacturing, hospitality or retail. This is true, but it is not a reason to dismiss the idea. In these sectors, work is tied to time, presence and staffing levels. Reducing working hours often requires more complex redesign, including changes to rotas, additional hiring or upfront investment. Colleagues and I have highlighted this when addressing the UK case of the NHS. But these challenges should be seen as design problems, not impossibilities. In fact, the potential benefits to society may be even greater in these sectors. Improved wellbeing and reduced burnout among healthcare staff and care workers can translate into better quality of service and fewer mistakes. Reduced working hours for healthcare staff could lead to fewer clinical mistakes. Iryna Inshyna/Shutterstock A more important concern is inequality. If working time reductions are adopted unevenly, there is a risk that some workers will be excluded – often those in lower-paid or frontline roles. This is a valid concern, but not an argument against the four-day week. Rather, it is an argument for implementing it more thoughtfully. Instead of asking whether all jobs can adopt the same model, the focus should be on how different forms of reduced work time can be adapted across sectors. This could include shorter daily hours, staggered schedules or phased time reductions. The future of work The renewed interest in reducing the amount of time we spend working is not happening in isolation. It is closely linked to broader debates about automation, productivity and the future of work. If technological advances continue to increase productivity, a fundamental question arises: who benefits from these gains? Historically – during the Great Depression, for example – working time reductions have been one way of redistributing those benefits. Compared with more radical proposals such as universal basic income, the four-day week offers a more direct and socially embedded way of sharing gains in productivity. The four-day week is not a universal solution, and it will not look the same everywhere. But the evidence shows working less can go hand-in-hand with maintaining productivity. It can also support a shift towards a society where time is valued not only as an economic input, but as a foundation for wellbeing, relationships and participation in community life. A century after the five-day week helped define modern work, there may be another turning point on the horizon. This time, the real question is not whether we can afford to reduce working time, but whether we can afford not to. Rita Fontinha’s employer, the University of Reading, has received funding from the Portuguese Government and the Azores Regional Government to conduct academic research on four-day working week pilots.

The first database built for AI Agents The post Ghost: A Database for Our Times? appeared first on Towards Data Science.

Discussion | Link

According to SAP, enterprise AI governance secures profit margins by replacing statistical guesses with deterministic control. Ask a consumer-grade model to count the words in a document, and it will often miss the mark by ten percent. Manos Raptopoulos, Global President of Customer Success Europe, APAC, Middle East & Africa at SAP, observes that the […] The post SAP: How enterprise AI governance secures profit margins appeared first on AI News.

Juli Clover / MacRumors: Tim Cook says that the Mac Studio and the Mac mini, used for AI and agentic tools, “may take several months to reach supply demand balance”  —  During today's earnings call for the second fiscal quarter of 2026, Apple CEO Tim Cook said that the Mac mini and Mac Studio could be hard to get for months to come.

Lauren Forristal / TechCrunch: Amazon launches “Join the chat”, an AI-powered feature that allows users to ask questions about products and receive conversational audio responses in real time  —  Amazon launched a new AI-powered feature on Tuesday that allows users to ask questions about products …

This guide explains the key differences between large language models and the broader...

Plus: Reddit and Spotify use authentic humans as a differentiator, 2,000 DESIGNmd templates, Deepmind’s Demis Hassabis on agents, AGI and more

Sycophancy eats away at truth and trust. Andriy Onufriyenko/Moment via Getty Images In the summer of 2025, OpenAI released ChatGPT 5 and removed its predecessor from the market. Many subscribers to the old model had become attached to its warm, enthusiastically agreeable tone and complained at the loss of their ingratiating robotic companion. Such was the scale of frustration that Sam Altman, OpenAI’s CEO, had to acknowledge that the rollout was botched, and the company reinstated access. Anyone who’s been told by a chatbot that their ideas are brilliant is familiar with artificial intelligence sycophancy: its tendency to tell users what they want to hear. Sometimes it’s very explicit – “that is such a deep question” – and sometimes it’s a lot more subtle. Consider an AI calling your idea for a paper “original,” even if many people have already written on the same topic, or insisting that your dumb idea for saving a tree in your garden still contains a germ of common sense. AI sycophancy seems harmless, maybe even cute, until you imagine someone consulting a chatbot about a weighty question, like a military strategy or a medical treatment. We study the impact of extensive human interactions with chatbots, and we recently published a paper on the ethics of AI sycophancy. We believe this tendency harms people’s ability to tell truth from fiction, and is psychologically and politically dangerous. Flattery over facts? In the simplest terms, sycophancy is the tendency to prioritize approval over factual accuracy, moral clarity, logical consistency or common sense. All AI models suffer from this trait, although there are some tonal differences between them. Open AI’s ChatGPT is often warm and affirming; Anthropic’s Claude tends to sound more reflective or philosophical when it agrees with you; and xAI’s Grok is insistently informal, even jocular. Politeness and adapting to someone’s communication style are not the same as sycophancy. Neither is using diplomatic language to convey sensitive information. A chatbot can be tactful without becoming sycophantic, just like a person can. Unlike people, though, AIs can’t be aware of their own sycophancy, because they are not – so far – aware of anything at all. Calling AIs sycophantic describes their patterns of behavior, not their character traits. The problem stems from the architecture of chatbot technology and the sources it draws from. Models are sycophantic because a great deal of language use on the internet – the raw material that chatbots learn from – displays sycophantic features. After all, humans often communicate with each other in sycophantic ways. Second, the training process to fine-tune AI models’ responses includes a kind of “quality control” carried out by human supervisors. This training method is known as “reinforcement learning from human feedback,” and it involves people rating chatbots’ comments for appropriateness and helpfulness. Human beings often are subject to an “agreeableness bias”: Our own preference toward sycophancy rubs off on models as we train them. Because of our own human bias for agreeableness, training can reinforce AI’s sycophancy. d3sign/Moment via Getty Images Finally, it’s hard to deny that sycophancy renders chatbots more likable. That, in turn, increases the chance that a given user will keep using it. It also increases the technology’s ability to extract user data, assuming that people are more likely to divulge information to a friendly bot. Truth and trust Why is this phenomenon so troubling? Let’s begin with AI sycophancy’s epistemic harms: how it hurts human users’ capacity to know the truth. The quality of any decision depends on a clear grasp of the facts pertaining to it. A general inquiring about the combat-readiness of an infantry division needs straightforward information. A CEO considering a merger with a competitor needs an honest assessment of the market conditions. A public health leader needs to know the real risk that an emerging pathogen poses. In all those cases, telling leaders what they might like to hear instead of the truth could lead them to make dangerous decisions. And the same is true in more humdrum contexts. People need to have the best information available before choosing a job, picking a major, buying a house or deciding on a medical procedure. In our February 2026 paper, we argue that sycophancy is also psychologically damaging. And that is true whether it comes from a person or from a chatbot. You never quite know if your very obliging interlocutor is being nice because they like you or because they want something. A shadow of suspicion creeps in: “Could my ideas really be that brilliant?” “Are my jokes really that hilarious?” This background music of doubt undermines the quality of the interaction. Sycophancy also undermines people’s capacity to know their own minds. If conversation partners – human or artificial – keep telling you how smart, funny and insightful you are, it damages your ability to identify your own weaknesses and blind spots. The psychological harms are compounded as people develop relationships with chatbots. The sycophancy of these models profoundly limits the kind of “friendship” you can have with them. In his classic account of friendship, Aristotle wrote that real friendship, which he calls a friendship of virtue, is based on trust and equality between the friends. You can’t trust a sycophant, because he doesn’t tell you the truth. And since he only tells you what you’d like to hear, he doesn’t put himself on an equal footing. AI conversations aren’t great prep for human ones. Natalia Lebedinskaia/Moment via Getty Images More importantly, interactions with sycophantic chatbots impart all the wrong habits for navigating the world of human relationships, where friction, disagreement, boredom and different opinions than your own are prevalent. AI sycophancy carries political risks as well. The success of liberal democracies has, traditionally, depended on the strength of their empirical and meritocratic mindset: on the ability of officials and citizens to identify, share and act on the truth. Historian Victor Davis Hansen famously attributed some of the Allies’ success in World War II to their ability to quickly recognize and address the faults of their strategic bombing campaigns. Lower-ranking officers were able to tell their superiors what wasn’t going well and argue forcefully for changing course. That was a real advantage over authoritarian competitors. Reining it in What can we do to reduce the risks? One promising approach is AI lab Anthropic’s embrace of what the company calls Constitutional AI: the attempt to teach chatbots to follow principles rather than mirror user preferences. But beyond technical innovations, it’s important to consider the policy side. One idea is to require AI companies to run and then publish sycophancy audits of their models – tests that show how well their products meet honesty benchmarks. We would argue that AI labs should also disclose sycophancy-related risks that emerge while training and testing their models, and the mitigation efforts they have undertaken. Some responsibility is on the users and their teachers: Schools and universities should be paying close attention to sycophancy as part of their AI literacy programs. But courts can also consider holding AI labs responsible for harms traceable to the sycophancy of their products, much as they are now contemplating social media companies’ responsibility for the addictive design of their platforms. As people interact more with chatbots, asking for advice about everything from whether your shoes go with your pants to how countries should conduct wars, the impact of AI’s sycophantic behavior is likely to become dramatic. Our intellectual, psychological and physical well-being requires taking this algorithmic vice very seriously. The Applied Ethics Center at UMass Boston receives funding from the Institute for Ethics and Emerging Technologies. Nir Eisikovits serves as the data ethics advisor to MindGuard, a startup focused on AI integration into companies' workflow. Cody Turner is a fellow at the Institute for Ethics and Emerging Technologies.

It takes a huge investment to be able to manufacture computer chips like these. Annabelle Chih/Getty Images The boom in data center construction is taking up much of the supply of high-tech components, especially processor and memory chips. This demand is squeezing consumer device makers, which are having trouble acquiring enough chips. This is happening even though data center servers and smartphones use different types of chips. The key distinction between consumer electronics and data centers is what they need chips to be optimized for. Smartphones and PCs require low power use, thermal efficiency and tight integration. Data centers that run AI systems such as large language models, or LLMs, require maximum compute power, memory bandwidth and storage throughput. To meet these needs, consumer devices tend to rely on systems-on-a-chip – chips that combine processing and storage – with dynamic random access memory, or DRAM, and NAND, a type of nonvolatile memory. In contrast, AI servers rely on graphics processing units, or GPUs, or other accelerator processors combined with high-bandwidth memory chips. I study global supply chains and how businesses respond to market constraints within these supply chains. The reason for the consumer electronics supply crunch has to do with the nature of the chip market: its concentration and high costs and how it responds to boom-and-bust cycles. AI is not replacing consumer electronics; it is reorganizing the chip market around new priorities for specific chip characteristics. Data centers are pulling capital and scarce memory capacity toward the production of accelerator processors and high-bandwidth memory and the data handling and electronics equipment that surround them. Chipmaking explained. A winner-takes-most industry Chip manufacturing behaves less like a competitive commodity market and more like a layered oligopoly. Scale matters because the leading firms can reinvest in research, improve yields, secure equipment and deepen customer relationships. In the case of graphics processor chips, designers such as NVIDIA, which has 85% market share, depend on advanced semiconductor foundries such as TSMC, which has more than 70% market share, to manufacture chips using extreme ultraviolet lithography machines from ASML, a monopoly. A small number of producers both design and manufacture memory chips. Currently, three companies – Samsung, Micron and SK Hynix – hold a majority market share in the memory chips market. Long development cycles, extremely high fixed costs and the need for technological leadership reinforce concentration over time. Consumer electronics firms such as Apple, along with other technology firms such as Amazon, Google, Microsoft and Xiaomi, increasingly design their own processor chips, because these chips shape the user experience, AI performance, power efficiency and system-level differentiation. Manufacturing memory chips, by contrast, is extraordinarily capital-intensive; requires high precision, efficiency and production line utilization; and is dominated by a few incumbent suppliers. Since 2000, the memory chip industry has moved through repeated cycles of overcapacity and undersupply: the post-dot-com collapse, the 2007-09 glut, the tighter 2010s after consolidation, the severe 2022-23 downturn, and the AI-driven tightness of 2024-25. This has led to high levels of concentration in the industry and chipmakers that are hesitant to add capacity. Producers often operate chip fabrication plants, or fabs, at or near capacity due to high fixed costs. The risk of having expensive facilities go underused keeps chipmakers from bringing new fabs online in lockstep with demand increases. Consolidation has reduced the number of major suppliers, who now increasingly direct investment toward higher-margin products rather than broadly adding capacity. That shift is important for understanding why AI demand is tightening chip supplies even as demand for consumer electronics continues to grow. The most advanced computer chips are made with a machine manufactured by one Dutch company. How the AI data center boom redirects capacity The AI boom has changed memory demand from a broad consumer cycle into a more segmented market centered on high-bandwidth memory chips. In 2023, Micron cut capital spending and the company’s fabs operated below levels needed to justify their cost. By 2026, however, Micron was reporting strong AI demand, record data center DRAM revenue and rapidly rising high-bandwidth memory sales. This shift matters because the market for supplying memory cannot respond quickly. Opening new fabs requires years of planning, large capital commitments and investments in advanced process equipment and skills. Memory chip manufacturers are likely to remain cautious about expanding capacity even as their profitability improves, with 2026 spending focused more on technology upgrades and high-value products than on large increases in chip supply. In practical terms, AI is not simply lifting all memory demand equally; it is redirecting scarce capacity toward massive, or hyperscale, data centers and server markets first. Can consumer electronics catch up? Consumer electronics can catch up, assuming the manufacturers can weather the cost increases from tariffs and geopolitical pressures. One way they could is by making investments to enable small AI language models to run on consumer devices, a move analysts expect the companies to attempt. Apple shifted a growing share of U.S.-bound iPhone production out of China to India and moved much of its iPad, Mac, Apple Watch and AirPods assembly for the U.S. market to Vietnam to lower the company’s tariff burden. Yet relocation does not eliminate cost pressure. Manufacturing iPhones in India still costs roughly 5% to 8% more than in China, and in some cases closer to 10%, because supplier ecosystems, logistics and production efficiency remain stronger in China. Rising geopolitical tensions between the United States and China led to supply constraints and export controls on critical minerals and chip components, raising input costs for consumer electronics manufacturers. This led to higher total import costs and reduced margins for firms unable to pass costs fully to consumers, leading to further consolidation in supply. Consumer devices do not need to replicate data center infrastructure to offer AI on their products. Their opportunity lies in running small language models on-device for summarization, rewriting, search, assistance and lightweight reasoning. Doing so, however, creates a distinct hardware requirement. Phones and laptops need to incorporate multiple functions on the same chip, combining processing capability with fast local memory and enough storage to keep on-device AI responsive. Apple’s current device requirements for the company’s AI, Apple Intelligence, also show that older phones often lack the compute power and memory needed for useful on-device AI. To adopt AI, device makers need to redesign their products with higher-end chips – both processors and memory – that can piggyback on the AI model-oriented growth in the chips market driven by the data center boom. Such a shift by the device makers could also provide a useful backstop for the memory chipmakers in case the projected AI and data center growth does not materialize in the medium to long term, a boom-and-bust cycle that memory chipmakers have had to endure many times in the past. Chipmakers have been devoting much of their precious manufacturing capacity to lucrative AI chips that are filling new data centers, like this Meta facility in Stanton Springs, Ga. AP Photo/Mike Stewart What this means for the wider economy The AI and data center boom is redistributing capital, supplier attention and pricing power across the broader economy. Sectors with limited purchasing leverage are especially vulnerable when chip supplies tighten. For example, medical technology accounts for less than 1% of the overall chip market, leaving essential equipment manufacturers exposed during shortages. In contrast, sectors linked to power delivery and digital infrastructure may benefit from the boom because they try to keep up with demand for cloud services and electrification. The International Energy Agency estimates that data centers consumed about 415 TWh of electricity in 2024 and notes that AI is accelerating the deployment of high-performance servers, which implies stronger demand for the grid, storage, cooling and networking equipment around them. For the consumer electronics industry, the strategic task is not to try to match the AI data centers chip for chip but to build differentiated, energy-efficient, on-device AI services while managing higher supply chain and tariff risks. And for consumers looking to buy phones, games and laptops, because of high demand from data centers, the next few years are likely to bring higher prices, shortages and delayed product releases. Vidya Mani has received funding from LMI. I am a Senior Research Fellow at the Inter-American Dialogue and a member of GRI

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. A new US phone network for Christians aims to block porn and gender-related content A new US-wide cell phone network marketed to Christians is set to launch next week. It blocks…

The public availability comes before the expected wider release of the vendor’s powerful and controversial Mythos cybersecurity model.

USE CASEEnterprise Risk Intelligence & Stress Testing Analytics Financial services...

Feeling creative? Have something to say about the last 20 years of cybersecurity? Our editors will award the best cybersecurity-related caption with a $20 gift card.

The company saw a 30% year-on-year jump in the number of people using search every week, CEO Steve Huffman said on Thursday.